Research On Cooperative Control Method Of UAV Formation

With the rapid development of science and technology and the advent of 5G communication era,more high-speed and reliable communication technology makes the cooperative control of UAV group gradually come into everyone’s view.Compared with single agent control,multi-agent cooperation has the advantages of wide application range and high flexibility.At present,it has a very broad application prospect in military,medical,logistics and other fields,In this paper,assuming that the communication diagram is undirected and strongly connected,the controller is designed for the quadrotor formation mission,and the stability of the controller in the case of time delay is analyzed,and the feasibility of the control algorithm is verified by simulation.The following is the main content of this paperFirstly,the mathematical model of the control object needed by the system is established,and the dynamic model and coordinate system of the quadrotor are established.On this basis,the establishment process of the simulation model of the quadrotor is introduced,the motor model of the quadrotor is abstracted,and the nonlinear constraints in the flight process of the quadrotor are analyzed.In addition,the quadrotor flight control is designed,and the simulation analysis is carried out for the fixed-point servo and the moving point servo respectively.The stability and rapidity of the flight control algorithm are verified from the simulation,which lays the simulation foundation for the multi four rotor formation simulation.Secondly,the mathematical description of multi-agent formation is introduced,and the basic knowledge of graph theory is briefly introduced.In addition,the consistency judgment method of formation control is briefly introduced.On this basis,the communication strategy between multi-agent is determined,and the communication strategy is abstracted as a strongly connected undirected graph.Under the condition that each agent can obtain the pose information of other agents,the cooperative control strategy is designed for the first-order agent model and the second-order agent model,and the sufficient conditions for the stability of the two control strategies are given.The simulation of servo task verifies the rapidity and stability of the control algorithm.Then,considering the collision avoidance problem between agents in the formation process,the design of artificial potential field is introduced,and repulsion field is set for each agent.When the distance is less than the threshold,the repulsion field behaves as repulsion,which can effectively avoid the collision between agents.But at the same time,it also introduces the local equilibrium point for the multi-agent system,that is,the agent maintains the equilibrium state when the formation target is not completed.In view of this situation,the traditional artificial potential field is improved and a small amount of interference is introduced into the artificial potential field to force the agent out of the local stable point,so as to achieve the goal of formation.The effectiveness of the above contents is verified by simulation.Finally,the communication delay in multi-agent system is analyzed,and the main delay of communication is transmission delay.It shows the design of cooperative control rate in the case of transmission delay.Based on the case that only transmission delay exists between multi-agent,the simulation of fixed delay and random delay is carried out,which verifies the stability of the cooperative control rate in the case of transmission delay within 100 ms.In addition,the simulation model of quadrotors is introduced,and the simulation analysis of multi quadrotor formation is carried out for the case of no delay,fixed delay and variable formation respectively.The feasibility of the cooperative control rate is verified from the simulation.